Macroeconomics of Green Energy*

Frederick R. Treyz, Mark D’Amato, Christine G. Crafton, Rod Motamedi** (fred@remi.com)

I.   Introduction

The recent summit in Copenhagen showcased the world’s focus on climate change in its most public and widespread stage to date.  While world leaders did not leave with a binding plan to reduce carbon dioxide emissions, they did agree on the basic form of the required measures.  In order to combat climate change the world must begin to use more alternative fuels and improve efficiency.  This requirement comes with many unknowns with respect to electricity prices, infrastructure investments, regulations, and their effects on the macro-economy.  We undertook this analysis in order to better understand some of the ramifications of these efforts.  Specifically, how changes in energy prices and investment would flow through a dynamic, structural macroeconomic model and how efficiency improvements would affect jobs, output, and regional competitiveness in the United States.

The sections that follow begin with an overview of the state of alternative energy measures in the United States.  The next section describes the structure of the PI+ model produced by Regional Economic Models, Inc.  This section also describes how various energy-related shocks flow through the model.  Section IV is a macroeconomic study of the recent weatherization program unveiled in the American Recovery and Reinvestment Act.  The next two sections are the conclusion and references.

II.   Overview of Green Energy in the US

Current Status of Green Energy Initiatives in the U.S.

Total U.S. investment in “green energy” equaled $23 billion in 2008 (DOE, 2009).  Green energy is often used synonymously with renewable energy or may include renewable energy with other energy-efficient technologies.  The improving economics of green energy development has fueled impressive growth over the past decade in both the demand for renewable energy and in the technologies to promote greater energy efficiency.  In 2008, 43% of all new electrical capacity in the U.S. was sourced from renewable plants, a large contrast to the 2% of new capacity additions only six years ago in 2004 (DOE, 2009).  Private investment in renewable technology also grew from $35 million in 2001 to $6 billion in 2008 (DOE, 2009).

Renewable energy technologies typically include solar thermal power plants and heating/cooling systems, wind farms, geothermal plants, hydroelectricity, solar photo-voltaics, ocean power systems and the use of biomass as the major components.  These technologies will become economically competitive with the fossil fuels that meet 85% of U.S. energy needs as rapid advancement in technology is driving down costs and increasing demand.  Hydroelectricity is currently the largest source of renewable power in the U.S. producing 34% of the total renewable power in 2008 (2.4% of the nation’s total electricity) (EIA, 2008).  Wood and wood-derived fuels constitute 28% of renewable power generation followed by biofuels (19%), wind (7%), waste (6%), geothermal (5%) and solar/PV (1%) (EIA, 2008).

The wind power industry is growing rapidly in the United States overall and experiencing significant growth in certain regions of the country.  Data from the American Wind Energy Association indicates that installed U.S. wind power capacity now exceeds 35 megawatts (MW), enough electricity to power 9.7 million American households (AWEA, 2009).  One quarter of the U.S. land area has winds strong enough to produce electricity at the same price as natural gas and coal (EESI, 2008) and will as a result foster higher wind power take up rates.  Texas leads the country in wind power development, followed by Iowa and California.  As of 2008, the U.S. leads the world in wind power generation and the U.S. Department of Energy (DOE) estimates that wind power could likely satisfy 20% of U.S. electricity demand by 2030 (DOE, 2009).

Based on a report issued by DOE's National Renewable Energy Laboratory (NREL), renewable-energy development is spreading rapidly throughout the country, following public policies designed to spur renewable energy growth.  In February 2010, the Obama Administration approved an $8 billion loan guarantee for the construction of two nuclear reactors in Georgia.  If the project goes forward, these would be the first plants built in the United States since the 1970s.  As of 2008, nuclear plants satisfied 9% of the nation's total electric energy consumption (EIA, 2008).  President Obama’s New Energy For America plan includes a federal investment of $150 billion over the next decade, with the goal of 10% of total electricity production sourced from renewable plants by 2012 and 25% by 2025.

According the EIA, commercial, residential and government buildings account for 39% of total energy use and 68% of total electricity consumption in the country.  Efforts are underway to both decrease consumption and promote efficiency.  As an example, the American Reinvestment and Recovery Act of 2009 awarded the Office of Energy Efficiency (EERE) $16.8 billion to improve efficiency through better building and industrial technologies, weatherization assistance, improved vehicle technologies, tax credits, grants, rebates, the construction of national laboratories, and other programs.

Specific State Results

According to the DOE NREL report, California led the nation in terms of total non-hydroelectric renewable generation in 2007, while Maine generated the largest percentage of electricity from renewable resources other than hydropower, at 26.1% (NREL, 2009).

A number of states have adopted the renewable portfolio standards (RPS), requiring that renewable energy sources be used to supply a certain fraction of retail electricity sales; and many of these states recently expanded their targets significantly. Today, 28 states plus the District of Columbia have RPS requirements, with renewable energy targets ranging from 2% to 40% of total electricity supply, to be achieved over the next five to 15 years.  At the end of 2007, these combined RPS policies, which cover 46% of the nation’s electricity load, called for utilities to procure about 16 million megawatt hours (MWh) of new renewable energy generation.  Going forward, they are expected to drive the development of more than 30,000 MW of new renewable energy capacity by 2015 if fully met (Wiser & Barbose, 2008).  All but a dozen states have implemented policies for connecting renewable-energy systems to the power grid, known as interconnection, while all but eight allow customers to earn credit for power fed back into the grid, a policy called net metering.

Table 1 Renewable Energy Target Production by State, Amount and Year


III.   The Structure of the REMI Model

The Economic Interactions of Energy Policy

In the most conventional sense, green energy is typically defined as electricity generated from non-polluting, environmentally-friendly, sustainable sources that harness the power of the wind, water, sun or earth.  In a somewhat broader view, this definition can be expanded to include any technologies or innovations that promote energy efficiency thereby reducing the amount of electricity produced from conventional sources like coal and natural gas.

However green energy is defined, there are a number of intuitive exogenous shocks to the macro-economy that can be modeled to approximate the comprehensive impact of a national policy shift towards clean energy technologies.  In general, whether the initiative is wind or weatherization, there will be some level of investment in the products and services that support the research and design, manufacturing, installation, and operation and maintenance phases of implementation.  Further, given that the policy has the desired effect, there will be some level of disinvestment in conventional energy generation and transmission technologies.  Finally, there will be a price effect that will have a direct impact on the end users of electricity.

Each of these impacts can be modeled using the REMI PI+ structural economic forecasting and policy analysis tool. PI+ integrates input-output, computable general equilibrium, econometric and economic geography methodologies.  Further, the model is dynamic, with forecasts and simulations generated on an annual basis and behavioral responses to compensation, price, and other economic factors.

The overall structure of the model can be summarized in five major blocks: (1) Output and Demand, (2) Labor and Capital Demand, (3) Population and Labor Supply, (4) Compensation, Prices and Costs and (5) Market Shares.  The blocks and their key interactions are shown in Figures 1 and 2 on the next page.

The Output and Demand block consists of output, demand, consumption, investment, government spending, exports and imports, as well as feedback from output changes due to the change in the productivity of intermediate inputs.  The Labor and Capital Demand block includes the determination of labor productivity, labor intensity, and the optimal capital stocks.  Labor force participation rate and migration equations are in the Population and Labor Supply block.  The Compensation, Prices, and Costs block includes composite prices, determinants of production costs, the consumption price deflator, housing prices, and the compensation equations.  The proportion of local, inter-regional, and export markets captured by each region is included in the Market Shares block.

As exogenous shocks are introduced into the model, the inherent linkages can be used to trace the ripple and feedback effects throughout the macro-economy that influence not only those sectors and regions directly affected by the policy shift, but all industries nationwide.  The following analysis traces the impact of an increase in investment for green energy sectors, a decrease in investment for conventional energy sectors, and an increase in the price of electricity in the short-term.  It should be noted that each of these general scenarios was considered in isolation with ceteris paribus assumptions.  The more complex analysis of a comprehensive policy shift that includes multiple simultaneous exogenous shocks is included in the case study that follows in the next section.

Figure 1.  REMI Model Linkages


Figure 2.  Economic Geography Linkages

Increase in Investment/Final Demand/Employment: Green Energy Sectors

Any move towards the more comprehensive use of green energy will be accompanied by some level of investment in labor or equipment needed to fully implement the outlined policies.  Within the model, these exogenous shocks can take the form of increased investment spending in producers’ durable equipment, increased final demand for relevant industries, or increased employment in relevant industrial sectors.  Ceteris paribus, these shocks would drive an increase in output at both the national and all regional levels.  It should be noted, however, that the effects will differ by region depending on the specific industry mix.

This increase in the overall strength of the economy will help stimulate job growth in industry sectors across the economy.  Those regions with industry mixes that leave them at a competitive advantage relative to other regions will experience economic in-migration, population gains, and an increase in labor force.  As the quality of the labor force increases, compensation rates will increase, which will in turn result in relatively higher costs of production.  As relative labor/production costs increase, it follows that imports will rise as exports fall.  Increased compensation rates, however, also lead to an increase in real disposable income, which is the primary determinant of consumption which in turn drives output.  Therefore, in this case, it is likely that the combined effect of an increase in investment and consumption would outweigh the net effect of export substitution and lead to an overall increase in gross domestic/regional product.

Decrease in Final Demand/Employment: Conventional Energy Sectors

Intuitively, a shift towards green energy will be accompanied by a shift away from conventional sources of electricity generation.  Within the model, those regions with a high concentration of conventional energy sectors will be the hardest hit.

In the same way that an increase in investment resulted in an overall net gain for the regional and national economy, a decrease in demand for industries that produce conventional energy will result in a net loss for the economy.  Those states like Wyoming and West Virginia will see the largest negative impacts and will experience significant out-migration, population loss, and labor force decline.  This will result in lower compensation rates which, while helping businesses lower their production costs, will also have a drastic negative impact on regional real disposable incomes.  This loss of purchasing power will lead to a reduction in consumption, which will lower regional output and employment across all industry sectors.

Change in Price: Electricity

It is generally accepted that a shift from conventional to green sources of electricity generation, at least in the short-term, will lead to higher electricity costs for both individuals and businesses.  Cost would increase because much of the infrastructure needed to support green energy would need to be put into place while conventional sources would only need to be maintained.  The question then becomes, at what point does the production of green electricity achieve the economies of scale to lower the per-kilowatt-hour price of electricity to end users?

To answer this, one must have a sense of the timeline needed to develop the domestic capacity for a number of different types of renewable energy technologies as well as an understanding of how these technologies will affect the quantity of electricity used over time.  This will effectively provide supply and demand curves that can be used to derive approximate price levels.

Within the REMI model, costs are defined as price times quantity.  Therefore, while costs may increase with a move towards green energy, efficiency may also increase thereby reducing the quantity demanded and resulting in ambiguous changes in total electricity costs.

In the interest of clarity, the following analysis will assume a ceteris paribus increase in the short-term price of electricity for both individuals and businesses (commercial and industrial).  From an individual consumer’s standpoint, we can imagine that this has the same negative effect on real disposable income that was seen with the disinvestment in the sectors associated with conventional energy production.  This effective loss of income will similarly reduce consumption and therefore output and employment economy wide.  From a commercial and industrial standpoint, this price change will greatly increase the cost of production which will effectively reduce exports and increase imports to the affected regions.  These combined negative effects to both individuals and businesses will influence economic migration patterns from the most affected areas to the least affected areas dramatically altering population and labor force characteristics nationwide.

Model Closures

The model uses three different methods to converge onto a solution.  Each has its own implications for the sustainability of employment and other macroeconomic gains.  These closures are described below.

Keynesian Response: The Keynesian closure allows economic policy shocks to have permanent effects on aggregate employment.  This closure is used when a monetary policy reaction is unwanted or unlikely.  In other words, this option will not use an interest rate mechanism to correct changes in U.S. employment that have been caused by an exogenous policy shock.

Anticipatory Cooperative Federal Reserve Response: The Anticipatory Fed closure assumes coordination between fiscal and monetary policy makers resulting in interest rate adjustments that would immediately adapt to new policies, so that employment is maintained at a constant rate.  In this case, there would be no (or brief) deviations in baseline employment in response to an exogenous shock.

This option uses a classical or rational expectations closure.  When chosen, an immediate interest rate mechanism is exerted to offset a shock introduced into the macro-economy.  The goal is to return U.S. unemployment back to a non-accelerating rate.

Historically Observed Response: The Historically Observed closure is a blend of the previous two closures. This closure assumes no monetary policy response during the first two-and-a-half years. After this time, there will be a response to return employment back to baseline, or non-accelerating levels.

This option combines the Keynesian response in the short term with the Anticipatory Cooperative Federal Reserve response thereafter.  The mechanism and goal are the same as in the Anticipatory Fed, but the timing is more gradual and allows for instances where monetary authorities either had no forewarning or allowed a wait-and-see period for the economy to correct itself before intervening.

IV.   Case Study

Introduction

As part of the American Recovery and Reinvestment Act (ARRA), the DOE received $5 billion to weatherize homes.  This program will focus on improving the homes of low income homeowners who typically cannot afford to undertake the effort themselves.  The money will be spent on updating or adding insulation, fixing drafty windows and doors, and other building modifications that would increase the energy efficiency of the home.

In this case study, we analyze the spending by state and the expected energy savings to ascertain what the ramifications of this program would be on the macro-economy of states and the nation.  With the expansion of climate and energy issues mentioned earlier, developing a better understanding of what these programs actually mean for the consumer and the economy is important to shaping the debate and finding efficient solutions.

Methodology

The central data in this study is the weatherization spending for each state and the District of Columbia (DC).  Using this data we were able to estimate the number of homes that would be weatherized and thus find the total amount of energy saved.  The following paragraphs outline the model, what data was used and how, and also the sources for that data.

We used a PI+ model of the United States which included 70 NAICS-based industries (66 private non-farm industries, state and local government, federal civilian, federal military, and farm) and 51 regions (all states and DC).  The dynamic features of PI+ were particularly attractive for this study as it allows for the cost and productivity changes that may result from the weatherization project.  The Keynesian closure was used for this study because a monetary policy response is highly unlikely in this case.  It would be counterproductive to the spirit of a government stimulus and unnecessary because the magnitude of the GDP change is not large enough to warrant such a response.

We obtained funding allocations by state from the DOE’s website for the Weatherization Assistance Program (DOE, 2009).[i] The spending is available by state and territory and by total allocation and funds awarded to date.  This study excludes allocations for U.S. territories, e.g. American Samoa and Guam, and for the Northern Arapaho Tribe and Navajo Nation because these regions were not available as separate regions within the model used.  The total allocation used in this study is roughly $4.67 billion.[ii]

The weatherization funding was modeled as an increase in final demand for construction, the industry that incorporates repairs and alterations to residential buildings.  Using the spending information from the DOE, we entered the amount awarded to date in the year 2009 and the remainder in 2010.  This is a valid assumption given that the money already awarded is almost exactly half the total allocation.  In order to account for the fact that some weatherization would have occurred without this program, we assumed one-fourth (0.25) free-ridership, which is conservative since the program is specifically targeting low-income homeowners (families making up to 200% of the federal poverty level) who likely would not be able to afford the repairs without this program.

After finding the total amount to be spent in each state, we then sought to find how many homes would be weatherized with that money.  According to the DOE, states will spend an average of $6,500 to weatherize each home (DOE, 2009).[iii] By dividing each state’s allocation by $6,500, we found the number of homes that will be weatherized.

The next step is to translate the number of weatherized homes into a dollar value of energy savings.  This task was a multistep process that began with finding average energy expenditures per household by state.  This process was done mainly using information from the EIA.  Average monthly residential electricity spending (EIA, 2010)[iv] was used to find annual electricity spending.  Residential consumption of natural gas (EIA, 2010)[v] (NG) multiplied by residential price of NG (EIA, 2010)[vi] divided by the number of residential customers for NG (EIA, 2010)[vii] was used to find annual NG spending.  Residential sales of fuel oil (EIA, 2009)[viii] multiplied by the price of fuel oil (EIA, 2009)[ix] divided by the number of households (U.S. Census Bureau)[x] was used to find annual fuel oil spending.  All the data was for the year 2008.

According to the DOE, each weatherized home can expect reductions of “32% for heating bills and savings of hundreds of dollars per year on overall energy bills (see footnote 2).” Because we examined electricity, natural gas, and fuel oil, which can all be used for heating, and if not used for heating would fall under the savings for other utilities, we chose to reduce expenditures on each category by 32%.  This assumption does result in somewhat larger savings than would otherwise occur but does not change the basic findings in the results.

The final step in finding total energy savings was to multiply the number of weatherized homes by 32% of the energy expenditures by state.  The energy savings entered the model as reductions in consumer spending on electricity, natural gas, and fuel oil and coal.  Furthermore, because low-income households and individuals typically have high marginal propensities to consume, we assumed that all energy savings would be spent on other consumption.  This money was entered using the consumption reallocation variable which assumes that the money would be spent proportionally across all consumption categories based on existing consumption patterns.  Lastly, to account for deterioration in the weatherization measures, we applied a straight-line depreciation of the energy savings starting on the eleventh year after installation and going to zero by the twentieth year.

Results

Summary

The results of the simulation are characterized by two distinct phases: the construction phase (2009 and 2010) and the savings phase (2011 through 2030).[xi] The construction phase shows the largest changes in the economy.  Table 2 summarizes the basic results by phase.

Table 2: Summary Results

Table 3: States with Negative Average Employment from 2011 - 2030 (Thousands)

In the construction phase, increased demand for construction directly creates an average of 16,789 construction sector jobs nationally in 2009 and 2010, which is 38% of total job creation over those two years.  As seen in the table above, the average job creation is significantly less during the non-construction years.  While positive at the national level, not all states see positive job creation during the savings phase.  Specifically, there are twelve states that have negative average employment during the savings phase, shown in Table 3.  While none of the job losses are high (for example Texas’ loss is around 0.001%), they appear predominately in energy and fossil fuel producing states like Texas, Oklahoma, Wyoming, and West Virginia.  California sees most of its job losses coming from construction and income-dependent industries like food services.  The non-energy producing states listed in Table 3 see their utilities employment fall but do not see employment increases in other sectors to offset the losses; meaning they capture little of the additional consumer spending within their borders.

The results for output and GDP show some interesting dynamics.  As Table 2 shows, national employment is positive for each phase and over all years, however the same is not true for output (not shown) and GDP.  The switch from consumer spending on fuel toward spending on other consumption has resulted in the creation of more jobs but with lower productivity and value-added.  It is this switch that allows for total employment to increase but the country’s net new economic activity to decrease.  This switch is also what allows employment to be positive in most states because the decrease in demand for utilities, and thus the fuels that power them, results in very few job losses due to those industries’ high productivity; a large change in output can be met with a relatively small number of employees.  On the other hand, the reallocation of consumer spending tends toward services, which are more labor intensive and thus create more direct jobs for a given amount of demand.

Nominal incomes rise along with employment over the simulation.  However, real incomes fall on average over the savings phase.  This fall is due to an increase in the PCE-Price index over the entire simulation phase.  The PCE-Price index increases due to an increase in production costs, which change in reaction to labor and capital costs.  These costs mainly increase during the construction phase in response to the sudden influx of dollars into the construction market.  These costs then decline over the savings period.  Figure 3 shows the relationship between real disposable income and the PCE index.

Figure 3: Percentage Change in Real Disposable Income and PCE-Price Index


Discussion

As mentioned in the methodology section, the energy savings were modeled as a reduction in consumer spending on electricity, natural gas, and fuel oil while an equal but opposite amount was used to increase all other consumer spending.  The decrease in spending results in a reduction in demand for companies in the utilities industry as it is these companies that provide electricity, gas, and fuel oil to households.  It is because of this demand reduction that all states see their utilities employment fall.  The largest average reductions in utilities employment over the savings period are seen in New York, at 22 jobs, and Texas, at 20.

The reduction in demand for utilities then has downstream effects to its main suppliers.  Significantly for the states mentioned in Table 3, oil and gas extraction and mining together account for roughly 18% of the value of output for the utilities sector over the savings phase.  Put another way, 20 cents of every dollar of utilities output goes to purchase the products of the oil and gas extraction and mining industries, especially oil and gas extraction which alone accounts for 14%.  This downstream effect determines why Texas is listed in Table 3 as having negative employment results, while New York is not.

The comparison of New York and Texas proves a useful way to probe the larger interactions and effects behind the results.  These two states are at opposite ends of the employment spectrum: New York gains the most jobs while Texas loses the most.  Each state’s experience highlights a different side of reallocating consumer spending away from paying utility bills toward other consumption.

As previously mentioned, New York loses the most utilities jobs, 22 versus Texas’ 20.  What keeps New York off the list in Table 3 is that New York also gains the most jobs during the savings period and it does so by capturing much of the additional consumer spending.  By being strong in services and income-sensitive industries, New York can offset its losses in the utilities sector by increasing employment in the newly-demanded services.  In New York, the top five industries with the largest consumption-supported employment gains over the savings phase are ambulatory health care, retail trade, private households, food services, and personal services.  The difference in Texas is that it loses too many high value-added, high income jobs from utilities and fossil fuel production that the incomes lost from those sectors is more than the employment in other sectors can provide.  Because of these combined losses from both the primary and secondary industries state personal income is negative while in New York it is positive.

Figure 4 shows total exports from each of the states to the rest of the US.  The figure does a good job of highlighting the different circumstances in each state.  The large growth in exports in New York during the savings phase contrasts sharply with the drop in exports in Texas.  In fact, the average gap is $17.95 million dollars per year with a cumulative gap of $359 million over 20 years.  Not surprisingly, Texas’ loss in exports mainly comes from the oil and gas extraction industry while New York’s biggest gain comes from the securities, investment, and commodities trading industry.  In other words, New York is better able to gain from the increase in consumer spending in other states, which creates a demand for financial services as businesses expand.

Figure 4: Exports to Rest of Nation

The same pattern of exports and imports and winners and losers carries over to other states as well.  In percentage terms, many of the states that see the largest employment gains are New England states that produce very little fossil fuels but many services while the states with the largest percentage losses are states like West Virginia, Louisiana, and Wyoming that depend on the production of fossil fuels.

V.   Conclusion

As the information presented in the first section of this paper shows, the rate of investment in renewable energy sources has far outpaced traditional energy sources in recent years.  This reality combined with the interactions described in the second section indicates a move toward a different energy infrastructure in the coming decades and the potential ramifications of such a change.  Because this infrastructure will remain in place for the foreseeable future, understanding it is vital for ensuring the health of the nation’s economy.

It is also important to remember the scale of the changes in this study.  New York’s job gains are only a difference of 0.0017% relative to the baseline and Texas’ job losses are only 0.0013%.  Such small changes will not spell disaster or windfall economic growth for any state, especially since in no case do they imply negative growth, only a slight decrease in hiring.

Scale aside, there are lessons to be taken from this analysis.  Firstly, if the nation takes a path toward reducing energy use, states must be prepared to reduce their reliance on fossil fuel production or capture more domestic market share while letting fuel demand reductions go to imports.  This import substitution may in fact result in both the benefits of reallocating consumer spending away from fuel and gains to the fuel producing states.  Secondly, this analysis does not capture the effects of making businesses more energy efficient.  Such gains would positively impact the country’s competitiveness through reductions in the cost of production and lower consumer prices causing real incomes to rise.  While increasing energy efficiency economy-wide would negatively affect fuel demand, any disadvantage to domestic producers could be offset for some time by import substitution.

VI.   Works Cited

AWEA. (2009). AWEA Year End 2009 Market Report. American Wind Energy Association.

DOE. (2009). 2008 Renewable Energy Data Book. U.S. Department of Energy.

DOE. (2009, June 26). DOE Weatherization Assistance Program: Obama Administration Delivers More than $304 Million for Weatherization Programs in Georgia, Illinois and New York. Retrieved March 31, 2010, from DOE Website: http://apps1.eere.energy.gov/weatherization/news_detail.cfm/news_id=12603

DOE. (2009, October 9). DOE Weatherization Assistance Program: Recovery Act Funding to the States. Retrieved March 31, 2010, from DOE Website: http://apps1.eere.energy.gov/weatherization/recovery_act_awards.cfm

EESI. (2008). Renewable Energy Becoming Cost Competitive with Fossil Fuels in the U.S. Enviromental and Energy Study Institute.

EIA. (2009, December 22). Adjusted Distillate Fuel Oil Sales for Residential Use. Retrieved March 31, 2010, from EIA Website: http://tonto.eia.doe.gov/dnav/pet/pet_cons_821dsta_a_EPD0_VAR_Mgal_a.htm

EIA. (2008). Annual Energy Review. Energy Information Administration.

EIA. (2010, January). Average Monthly Bill by Census Division, and State. Retrieved March 31, 2010, from EIA Website: http://www.eia.doe.gov/cneaf/electricity/esr/table5.html

EIA. (2010, March 29). Average Residential Price. Retrieved March 31, 2010, from EIA Website: http://tonto.eia.doe.gov/dnav/ng/ng_pri_sum_a_EPG0_PRS_DMcf_a.htm

EIA. (2010, March 29). Number of Natural Gas Residential Customers. Retrieved March 31, 2010, from EIA Website: http://tonto.eia.doe.gov/dnav/ng/ng_cons_num_a_EPG0_VN3_Count_a.htm

EIA. (2010, March 29). Residential Consumption of Natural Gas (Summary). Retrieved March 31, 2010, from EIA Website: http://tonto.eia.doe.gov/dnav/ng/ng_sum_lsum_a_EPG0_vrs_mmcf_a.htm

EIA. (2009, August 26). U.S. Total Refiner Petroleum Product Prices. Retrieved March 31, 2010, from EIA Website: http://tonto.eia.doe.gov/dnav/pet/pet_pri_refoth_dcu_nus_a.htm

NREL. (2009). State of the States 2009: Renewable Energy Development and Role of Policy. U.S. Department of Energy - National Renewable Energy Laboratory.

U.S. Census Bureau. (n.d.). Custom Table. Retrieved March 31, 2010, from Census Bureau Website: http://factfinder.census.gov/servlet/CTTable?_bm=y&-context=ct&-ds_name=ACS_2008_1YR_G00_&-mt_name=ACS_2008_1YR_G2000_B11012&-tree_id=308&-geo_id=04000US01&-geo_id=04000US02&-geo_id=04000US04&-geo_id=04000US05&-geo_id=04000US06&-geo_id=04000US08&-geo_id=0

Wiser, R., & Barbose, G. (2008). Renewables Portfolio Standards in the United States. Berkeley, CA: Lawrence Berkeley National Laboratory.

 

* This article copyrighted and reprinted by permission from the International Association for Energy Economics.  The material first appeared in the online proceedings of the 33rd IAEE International Conference, Rio de Janeiro, June, 2010.

** Frederick R. Treyz, Mark D’Amato and Rod Motamedi are with Regional Economic Models, Inc.; and Christine G. Crafton is with Booz Allen Hamilton. Contact: Frederick R. Treyz, CEO, REMI, 433 West St Amherst, MA 01002 USA, E-Mail: fred@remi.com, Phone: +1-413-549-1169 x 233.

 


 

[i] http://apps1.eere.energy.gov/weatherization/recovery_act_awards.cfm

[ii] This equals the $5 billion total program budget less funding reserved for technical assistance, DOE operations, and the excluded territories.

[iii] http://apps1.eere.energy.gov/weatherization/news_detail.cfm/news_id=12603

[iv] http://www.eia.doe.gov/cneaf/electricity/esr/table5.html

[v] http://tonto.eia.doe.gov/dnav/ng/ng_sum_lsum_a_EPG0_vrs_mmcf_a.htm

[vi] http://tonto.eia.doe.gov/dnav/ng/ng_pri_sum_a_EPG0_PRS_DMcf_a.htm

[vii] http://tonto.eia.doe.gov/dnav/ng/ng_cons_num_a_EPG0_VN3_Count_a.htm

[viii] http://tonto.eia.doe.gov/dnav/pet/pet_cons_821dsta_a_EPD0_VAR_Mgal_a.htm

[ix] http://tonto.eia.doe.gov/dnav/pet/pet_pri_refoth_dcu_nus_a.htm

[x] Number of households from the American Community Survey from the Census Bureau. Last accessed March 31, 2010.

[xi] All results presented in this section are in terms of absolute differences relative to the baseline, or “business as usual” scenario.

Click to view a printable version of this article.